论文阅读：Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models-CSDN博客

本文链接：https://blog.csdn.net/zyx302626/article/details/147375739

1. 背景&动机：

Recent advancements in reasoning with large language models (RLLMs), such as OpenAI-O1 and DeepSeek-R1.A central factor in their success lies in the application of long chain-of-thought (Long CoT) characteristics, which enhance reasoning abilities and enable the solution of intricate problems. However, despite these developments, a comprehensive survey on Long CoT is still lacking, limiting our understanding of its distinctions from traditional short chain-of-thought (Short CoT) and complicating ongoing debates on issues like "overthinking" and "test-time scaling".

2. 怎么做的/做了些什么：

2.1. Systematic Distinction

Introduce the concept of Long CoT reasoning and distinguish it from the traditional Short CoT, thereby providing a clear framework for understanding both paradigms and their respective characteristics.

区分长短链推理的关键点：

(1) Deep Reasoning

(2) Extensive Exploration

(3) Feasible Reflection